Type=Integer(+)

Default = 1000

Maximum number of iterations the program will make through the optimization loop. Typically, the Analyst Drive program will converge when the cost function value no longer decreases from one iteration to the next, or when the search step length α becomes smaller than the TOL parameter value. For some problems (particularly dynamic ones) the solution will just continue decreasing in value indefinitely even if the decrease is small, and these conditions may not be met for a substantial amount of time. For these scenarios it is recommended to use the maximum iterations value to control the estimation process by running for a fixed number of iterations, and then sending the updated OD matrix back through the assignment program Highway or Avenue. This strategy will produce good results, however, setting the maximum number of iterations to a very high value and allowing the program to converge will yield the greatest possible reduction in the cost function for a single run but may be less efficient than a strict bi-level scheme. This should be done if re-assignment is not desired due to excessively long run times, or if the user feels it is simply not necessary (e.g. a single run through the process is all that is required).